PMI · free study guide

PMI-ACP study guide

Agile Certified Practitioner

Free PMI-ACP study guide and exam blueprint. 120 questions in 180 minutes. Written from PMI's ECO Nov 2024, last checked 15 July 2026. Domain weightings, glossary, and practice questions with worked explanations.

Written from PMI's ECO Nov 2024. Last checked against that document on 15 July 2026.
120 questions 180 minutes 4 domains
Orientation

Not a Scrum exam.

The most expensive assumption you can bring to PMI-ACP is that it tests Scrum. It tests whether you can pick an approach, read a situation, and move value through a system, across Scrum, Kanban, Lean, XP and scaled settings alike.

If you hold a PMP
PMI's eligibility table lists an active PMP as a route in by itself: it satisfies the agile and hybrid experience requirement outright. You still need 21 hours of formal agile training, but the experience hurdle is gone. If you passed PMP recently, this is the cheapest second credential available to you.

What you are walking into

120 items in three hours. Only 100 are scored; the other 20 are unscored pretest items scattered at random, which is PMI gathering difficulty data on questions before they count. You cannot tell which is which, so treat all 120 as real.

There is one 10-minute break, after item 60. PMI is explicit that once you review your answers and start that break, you cannot return to the first 60 items. That single sentence should shape how you pace the first half.

PMI does not publish a passing score, and nobody outside PMI knows it. Ignore anyone who tells you the number.

Getting in

The order that works

Your study plan

Exam Content Outline

What the exam is made of.

Four domains, 24 tasks. PMI publishes the weightings, and they are the first thing that surprises people: the two biggest domains are the two least concrete.

28%Mindset25%Leadership19%Product28%Delivery
Four domains · 24 tasks 100 scored items

Mindset and Delivery carry 56% between them. Product, the domain that looks most like a job description, is the smallest at 19%. Candidates who prepare by revising backlog mechanics are optimizing for the smallest slice of the paper.

The four domains

DomainWeightWhat it covers
I. Mindset28%Experiment early, agile mindset and complexity theory, collaborative team environment, transparency, psychological safety, shorter feedback loops, embracing change. Seven tasks.
II. Leadership25%Empower teams, facilitate problem resolution, promote knowledge sharing, promote agile principles, promote shared vision, facilitate conflict management. Six tasks.
III. Product19%Refine the backlog, manage increments, visualize work, manage value delivery. Four tasks.
IV. Delivery28%Seek early feedback, manage agile metrics, manage impediments and risk, eliminate waste, continuous improvement, engage customers, optimize flow. Seven tasks.

The format

Items120 total: 100 scored, 20 unscored pretest, randomly distributed and unmarked.
Time3 hours. Roughly 90 seconds per item.
TypesMultiple choice, multiple response, drag-and-drop, and exhibits.
BreakOne, after item 60. Once you start it you cannot return to items 1 to 60.
Pass markNot published by PMI.
DeliveryPearson VUE test center or online proctored.
RetakeThree attempts per one-year eligibility period, then a one-year wait.
Renewal30 PDUs every three years.
Read this
PMI describes the modernized ACP as capturing Enterprise Agility, and the outline says so plainly: the exam assesses not only team-level practice but whether you can align work to organizational priorities, optimize end-to-end flow, coordinate across teams, and engage stakeholders for outcomes. If your agile experience stops at one team's board, that sentence is the gap you are closing.

What this exam is not

The part nothing else prepares you for

Complexity and suitability.

This sits inside Mindset, the largest domain, and it is the content a Scrum background does not touch. PMI names three complexity models by name and expects you to classify a scenario with them.

Why complexity is on an agile exam

Agile is not universally correct, and PMI is explicit that you should be able to interpret the output of agile suitability tools and integrate models based on need. The underlying claim is that the right approach depends on the kind of problem you have. Complexity models are how you name the kind of problem, and they are what turns "should we be agile?" from an argument into an assessment.

Cynefin

Five domains, each with a different decision approach. Learn the sequence for each, because the exam tests whether you know that clear and complex demand opposite behavior.

DomainApproach
Clear (obvious)Sense, categorize, respond. Known best practice exists. Apply it. Do not experiment; you are wasting money rediscovering the answer.
ComplicatedSense, analyze, respond. Good practice exists but needs expertise. Bring in the expert. Analysis pays off here.
ComplexProbe, sense, respond. Cause and effect are only clear in hindsight. Run safe-to-fail experiments. This is where agile earns its keep.
ChaoticAct, sense, respond. No time to analyze. Act to establish order first, then work out what happened.
Confused (disorder)You do not know which domain you are in. The most dangerous place, because people default to whatever approach they personally prefer.

The exam pattern: a stem describes a situation and asks what to do. Classify it first. A well-understood, repeatable problem does not need a sprint; an unprecedented one does not need a detailed plan.

The Stacey Matrix

Two axes: how much agreement there is about what is wanted, and how much certainty there is about how to do it. Close to agreement and close to certainty is simple, and suits a predictive approach. Far from both is chaos. The large middle region, where requirements are contested and technology is unproven, is the complex zone where iterative and adaptive approaches fit.

The practical use is diagnostic. A project failing with agile is often close to agreement and certainty, where the ceremony is pure overhead. A project failing with a plan is usually far from both.

Complex adaptive systems

Agile suitability

The outline asks you to interpret the output of agile suitability tools. These assess a project across dimensions and produce a picture rather than a verdict.

Tell
When a stem gives you an organization pushing back on agile, the weak answer argues for agile. The strong answer assesses suitability, names the specific impediment, and addresses it. PMI is not testing whether you are an advocate. It is testing whether you can tell what kind of problem you are looking at.
Domain I 28% · 7 tasks

Mindset.

The largest domain, tied with Delivery. It is about the conditions that make agile work rather than the practices themselves, which is why it is hard to revise and easy to underestimate.

The seven tasks

  • Build an increment to validate the solution or the market need. The increment is a probe, not just a delivery.
  • Create an environment to innovate, learn and grow.
  • This is the complex domain of Cynefin in practice: when cause and effect are unclear, you learn by probing rather than by analyzing harder.
  • Design thinking and lean startup both appear in the outline as ways to shorten the loop between building and learning.
  • Use agile values and principles. The manifesto's shape matters: the left side is valued more, the right side still has value.
  • Apply complexity theory to classify a scenario, using complex adaptive systems, the Stacey Matrix, or Cynefin.
  • Identify the application, importance or risks of each complexity model given a scenario.
  • Interpret the output of agile suitability tools.
  • Integrate agile models based on need, at the level of a use case, a team, a project or an organization.

This task is why the complexity section exists. It is the only place on any PMI exam where you are asked to name a complexity model, and it sits in the biggest domain.

  • Establish team vision and working agreements. Written by the team; that is what makes them hold.
  • Form and develop a high-performing team.
  • Use retrospective findings to improve the team, not merely record them.
  • Use collaboration practices to break down silos.
  • Commit to the team's decisions even in disagreement. Disagree and commit is named explicitly.
  • Evaluate the team's understanding of agile and tailor the approach to it.
  • Identify the factors that decide the inter-team coordination approach, such as scrum of scrums or team of teams.
  • Make status, progress, process, risks, impediments and learning accessible to everyone, typically through information radiators.
  • Establish a feedback loop for the team.
  • Define communication strategies for co-located and distributed teams. Osmotic communication is free when co-located and must be engineered when not.
  • Note what is on the list: risks and impediments are radiated too, not just task status.
  • Promote a no-blame culture by encouraging objectivity. Blame makes information expensive, and empiricism runs on information.
  • Encourage dialogue over debate. Debate has a winner; dialogue has an outcome.
  • Solicit and provide constructive feedback, and act on it. The acting is the part that makes people offer it again.
  • Encourage challenging the status quo.
  • Psychological safety is not comfort. Safe teams argue more, because the argument is about the work rather than about status.
  • Include stakeholders from day one.
  • Maximize value within a given timeframe.
  • Use techniques that shorten feedback, such as design thinking and lean startup.
  • The shorter the loop, the cheaper being wrong is. That is the whole economic argument for iteration.
  • Promote a growth mindset in responding to change.
  • Embrace process adaptation by responding to changing requirements and priorities.
  • Encourage and model cross skills, described in the outline as generalizing specialists: deep in one area, capable across several.
  • Adapt to product needs based on learning and feedback.

Generalizing specialists matter because a team of pure specialists creates queues at every handoff, which is a flow problem wearing a staffing costume.

Domain II 25% · 6 tasks

Leadership.

A quarter of the exam, and almost all of it is influence without authority. If you hold a PMP, most of this will feel familiar; the vocabulary is agile but the judgment is the same.

The six tasks

  • Establish an environment of trust through transparent communication.
  • Motivate team members to experiment and take risks.
  • Coach and mentor team members.
  • Promote collective ownership of goals.
  • Recognize the difference between training, coaching and mentoring, and when each applies. Training gives knowledge. Coaching draws capability out for a specific goal. Mentoring shares experience over a longer relationship.
  • Apply emotional intelligence to support the team, build empathy, resolve conflict and influence positively.
  • Interpret non-verbal cues during team interaction.
  • Interpret the output of self-assessment tools to help teams develop.
  • Investigate the root cause, using root cause analysis or an Ishikawa diagram.
  • Determine resolution strategies with the team that add the most value. With, not for.
  • Ensure problems are resolved in a timely manner.
  • The order is the answer to most questions here: understand the cause, involve the team in the fix, then make sure it actually happens.
  • Create an environment to capture and share knowledge: lessons learned, retrospectives, communities of practice.
  • Leverage organizational knowledge assets from similar initiatives, people and processes.
  • Allocate time for knowledge sharing and for making the updates. Knowledge sharing that is not in the plan loses to the work every time.
  • Create awareness of agile values and principles.
  • Foster an environment for continuous improvement.
  • Recognize, celebrate and encourage agile behavior. Reinforcement is a leadership act, not an HR one.
  • Define and ensure common understanding of purpose and vision with all stakeholders. Common understanding, not just a published statement.
  • Ensure the product is always aligned to the vision and organizational goals. This is the Enterprise Agility thread.
  • Continuously communicate the vision and purpose. Once at kickoff is not communication.
  • Identify the root cause and the level of the conflict. The level matters: a problem to solve is not the same as a fight.
  • Promote a collaborative approach to resolving it.
  • The leader facilitates the resolution; they do not impose it. An imposed outcome is complied with and returns.
Domain III 19% · 4 tasks

Product.

The smallest domain, and the one candidates over-prepare because it looks most like the job. Four tasks, 19% of the paper.

The four tasks

  • Clarify the backlog items.
  • Prioritize with the customer or stakeholder. With them, which is the collaboration the outline keeps insisting on.
  • Decompose items as needed.
  • Use tools and techniques to collectively size work. Collectively: the people doing the work size it, together.

Sizing techniques named in the outline: relative sizing, story points, T-shirt sizing, wideband Delphi, planning poker, affinity estimating, ideal time.

  • Ensure the increment aligns with business priorities.
  • Define the increment goals.
  • Demonstrate increments of value for early feedback.
  • Measure the delivery of value, not the delivery of output.
  • Educate the team and stakeholders in work visualization techniques. Teaching is part of the task.
  • Establish a process to keep the data current. A stale board is worse than none, because it is trusted.
  • Continuously share the information.
  • Tools: Kanban board, task board, burndown and burnup charts, cumulative flow diagrams, information radiators.
  • Define what value looks like. The outline names the dimensions explicitly: success criteria, sustainability, security, privacy, regulatory, compliance.
  • Ensure value increments are optimized.
  • Ensure targeted results are achieved, such as customer satisfaction or increased sales.

Note what PMI put in the definition of value: security, privacy, regulatory and compliance. Value is not only features a customer asked for, and an answer that trades compliance for speed is wrong here as surely as it is on PMP.

Prioritization tools named: ROI, NPV, IRR, compliance, customer-valued prioritization, requirements reviews, MVP, minimal marketable feature, relative ranking, MoSCoW, Kano.

Domain IV 28% · 7 tasks

Delivery.

Tied with Mindset as the largest domain. This is flow, metrics, waste and risk, and it is where Lean and Kanban carry more weight than Scrum.

The seven tasks

  • Evaluate customer satisfaction.
  • Deliver work in small increments. Small batches shorten the loop and shrink the blast radius.
  • Collect and incorporate stakeholder feedback regularly.
  • Determine which metrics are appropriate for a given audience. A team metric is not an executive metric.
  • Radiate metrics to the relevant audience.
  • Review and analyze them, and use the insight for decisions. A metric nobody decides anything with is decoration.

Metrics named in the outline: velocity, throughput, productivity, cycle time, lead time, EVM for agile projects, defect rates, approved iterations, work in progress.

  • Proactively identify risks and impediments.
  • Engage the team to find the most appropriate course of action.
  • Prioritize impediment removal and risk mitigation.
  • Monitor and control both.
  • Use lessons learned to stop them recurring.

Agile risk tools named: risk-adjusted backlog, risk burndown graphs, risk-based spike, architectural spike.

  • Visualize the end-to-end flow of value, separating value-added from non-value-added. This is value stream mapping.
  • Use metrics, tools and feedback loops to identify waste.
  • Prioritize waste reduction.
  • Iterate on identifying and reducing it. It is a cadence, not a project.
  • Obtain metrics and feedback to drive improvement.
  • Implement improvement actions. Implement, not identify.
  • Evaluate the effectiveness of the improvement. The loop closes or it was theater.

Process improvement tools named: kaizen, the five whys, retrospectives and intraspectives, process tailoring and hybrid models, value stream mapping, control limits, pre-mortem, fishbone analysis.

  • Identify and analyze customers and their needs.
  • Validate that iteration deliverables meet acceptance criteria.
  • Encourage collaboration between customer and team, rather than acting as the channel between them.
  • Limit work in progress at all levels. At all levels: not just the team board.
  • Shield the team from interruptions, for example by creating team interfaces.
  • Use metrics to analyze and improve flow.

Little's Law is the arithmetic: cycle time = WIP ÷ throughput. Halve the WIP and cycle time halves without anyone working faster. It is why the answer to a jammed board is to lower the limit and swarm the constraint, never to start more work.

Tell
Delivery is where the exam stops being Scrum-shaped. Flow, waste, WIP, cycle time, value streams and the theory of constraints are Lean and Kanban ideas, and they carry 28% of the paper. A candidate who only knows sprints will feel this domain go past them.
Reference

Glossary.

Terms as the ACP outline uses them, weighted toward the ones a Scrum background does not supply.

Affinity estimating
Grouping many items by relative size quickly, without discussing each in depth. Suits a large backlog where per-item precision is not yet worth the time.
Complex adaptive system
Many interacting agents whose collective behavior is not predictable from the parts. Exhibits emergence, non-linearity, self-organization and feedback loops.
Cynefin
Five domains with different decision approaches: clear (sense-categorize-respond), complicated (sense-analyze-respond), complex (probe-sense-respond), chaotic (act-sense-respond), confused.
Design thinking
Empathize, define, ideate, prototype, test. Named in the outline as a way to shorten feedback loops.
Disagree and commit
Committing to the team's decision even when you argued against it. Named explicitly under task 1.3.
Enterprise Agility
PMI's framing for the modernized ACP: aligning work to organizational priorities, optimizing end-to-end flow, coordinating across teams, adapting at scale without losing coherence.
Generalizing specialist
Deep in one area, capable across several. Reduces the queues that form at every handoff in a team of pure specialists.
Ideal time
Effort assuming no interruptions. Named as an estimation approach; it invites comparison with calendar time, which is its weakness.
Intraspective
Named alongside retrospectives in the outline. An inward look at one's own practice rather than the team's process.
Lean startup
Build-measure-learn with validated learning. Named as a feedback-shortening technique.
Little's Law
Cycle time = WIP ÷ throughput. The arithmetic behind WIP limits.
Minimal marketable feature
The smallest feature that delivers market value on its own. Named alongside MVP in value-based prioritization.
Osmotic communication
Information absorbed by proximity. Free when co-located; must be engineered deliberately when distributed.
Pre-mortem
Imagine the project has failed, then work backward to why. Surfaces risks that optimism suppresses.
Risk-adjusted backlog
A backlog ordered accounting for risk as well as value, so risky items are addressed early enough to matter.
Risk burndown graph
Tracks total risk exposure over time, trending down as risks are retired.
Spike
A timeboxed investigation to reduce uncertainty. Risk-based spikes probe a risk; architectural spikes probe a technical approach.
Stacey Matrix
Two axes: agreement about what, certainty about how. Close to both is simple and suits prediction. Far from both is chaos. The middle is complex and suits adaptive approaches.
Value stream mapping
Mapping end-to-end flow and separating value-added from non-value-added steps. Usually reveals that work spends most of its life waiting.
Wideband Delphi
Iterative anonymous estimation with feedback between rounds until estimates converge. Removes the anchoring effect of the loudest voice.
Common questions

PMI-ACP exam questions, answered

How many questions are on the PMI-ACP exam?

The PMI-ACP exam has 120 questions and lasts 180 minutes. PMI uses a scaled score rather than a published percentage, so there is no fixed pass mark to quote.

What is on the PMI-ACP exam?

The exam covers 4 domains: Mindset (28%), Leadership (25%), Product (19%), Delivery (28%). These weightings come from PMI's ECO Nov 2024.

Is this PMI-ACP material current?

Yes. This guide was written from PMI's own published outline, ECO Nov 2024, and last checked against that document on 15 July 2026. That date is shown on every page so you can judge for yourself rather than take our word for it.

Are there free PMI-ACP practice questions?

Yes. Sample questions with full worked explanations are free and need no account. The complete question bank and the full-length 120-question timed mock are paid.

Practise it

The guide above is free. The question bank and the full-length timed mock are the paid part.

Open PMI-ACP →
Other exams:
PMP · CAPM · PSM I · PRINCE2 · ITIL · ISTQB CTFL · CC · Network+ · Security+ · CISSP · AWS CCP

Cutscores is independent and is not affiliated with, authorised by, endorsed by or sponsored by PMI or any certification body. PMI-ACP and all certification names and marks are the property of their respective owners, and are used here only to describe which exam this material prepares you for. We do not reproduce live exam content.